The long term goal of this work is to understand the biology of the tissue factor pathway inhibitor (TFPI) in health and disease states. More specifically, the following hypotheses will be tested, 1) That lipoprotein associated TFPI is structurally different from heparin releasable TFPI due to either, a) differential glycosylation, or b) carboxyl terminus proteolysis of lipoprotein associated TFPI. In addition, that this structural difference determines the functional differences between the two molecules, i.e., different binding affinities to lipoproteins and glycosaminoglycans. Characterization of the two purified molecules will be carried out using a variety of protein and carbohydrate biochemical techniques. In addition, the two molecules will be characterized by functional assays for factor Xa and factor VIIa/tissue factor inhibition, heparin binding, and lipoprotein binding. 2) That endothelial cells make the great majority of intravascular TFPI in vivo and that a pathway exists transferring TFPI from the endothelial cell surface to plasma lipoproteins. Combined biochemical, physiological, and immunohistochemical studies will be done to localize the source and pathway of metabolism of TFPI in vivo. 3) That the endothelial cell TFPI pool, as measured by levels of heparin releasable TFPI, rather than the lipoprotein associated TFPI pool, as measured by levels of plasma TFPI, is an indicator of thrombotic risk. An immunoassay that has been used to measure levels of plasma TFPI will be used to measure levels of TFPI before and after intravenous heparin in controls and in patients with thrombosis. 4) That high density lipoprotein associated TFPI is in part responsible for the inverse correlation between HDL levels and atherosclerosis. Fractionation of HDL will be done to determine if TFPI is present in HDL fractions that are atheroprotective. If initial studies suggest a relationship, further immunohistochemical and physiologic studies will follow.